246 Mr. A. T. Mukerjee on a Method of Measuring 



system- can be varied by the sliding condenser, the capacity of the 

 electroscope can be measured at different voltages. There is a 

 small but definite increase of capacity with voltage. 



The method is compared with those of Lester Cooke (Phil. Mag. 

 vol. vi. p. 410, 1903) and T. Barratt (Proc. Phys. Soc. London, 

 vol. xxviii. pp. 162-171, 1916). The capacity of a small p-ray 

 electroscope charged to about 100 volts has been measured by the 

 three methods with the following results : — Present method 

 0*78 Hh *01 : Lester Cooke's method, as improved by the author, 

 0-75 ± -01 ; and Barratt's method 073 ± -03 E. S. tl. 



rr^HE modern gold-leaf electroscope is extensively used by 

 _l_ workers in radio-activity and atmospheric electricity. 

 Many cases naturally arise where a knowledge of its capacity 

 becomes necessary. The capacities of such electroscopes, 

 being extremely small, of the order of one micro-microfarad, 

 cannot be determined by the usual methods. Ordinary 

 commutator methods lose their sensitiveness for such small 

 capacities. Even the ordinary method of mixtures cannot 

 be employed with a Dolezalek electrometer, as the probable 

 error of a single observation is usually of the same order as 

 thp quantities to be measured. 



The first suggestion for measuring such small capacities 

 was made by Borgmann * and Petrowsky f in 1899, but the 

 method was complicated and not very satisfactory in practice. 

 C. T. R. Wilson J measured the capacity of an Exner's electro- 

 scope in 1901 by charging it and determining the fall of 

 potential resulting from contact with a brass ball suspended 

 by a silk fibre. This method, again, can only be regarded as 

 an approximation, and Harms § mentions that individual 

 values obtained by this method differ by as much as 25 per cent. 

 Several other and much less simple methods have been 

 described by Harms || (1904) and Lichtenecker 1(1912), but 

 none are sufficiently accurate to justify a detailed discussion. 



The best methods hitherto used are those of Lester Cooke ** 

 (1903) and T. Barratt ft (1916). I have used both these 

 methods myself, and the results, which are detailed later on, 

 show that Lester Cooke's method is quite good, provided 

 that satisfactory arrangements are made for insulating and 



* Phys. Zeit. \ol. ii. pp. 651-653 (1901). 



+ Cumptes Rendus, vol. cxxviii. pp. 420-422 (1899). 



+ Roy. Soc. Proc. vol. Ixviii. p. 157 (1901). 



§ Phys. Zeit. vol. v. p. 47 (1904) & Ann. d.Plujs. vol. x. p. 816 (1903). 



|| Loc. cit. 



H Phys. Zeit. vol. xiii. pp. 516-518(1912). 



** Phil. Mag. vol. vi. p. 410 (1903). 



tf Proc. Phys. Soc. London, vol. xxviii. pp. 162-171 (1916). 



